|
Vintage Amateur and Military Radio Amateur/military receivers and transmitters, morse, and any other related vintage comms equipment. |
|
Thread Tools |
30th Oct 2017, 6:33 pm | #1 |
Pentode
Join Date: Aug 2010
Location: Isle of Wight, UK.
Posts: 217
|
Racal RA17W.
Hello all
does anyone know what the 'w' suffix indicates in the RA17 [W] nomenclature? Serial number is 9051. Are there any significant circuit or specification variations? I have restored one RA17L which is now performing admirably. I was lucky because apparently the 40Mhz filter had not been interfered with. I am about to embark on the 'W' and another 'L' and I suspect my 40Mhz luck might 'run out'! So, what might be the most cost-effective way of checking the 40mhz filter without acquiring a wobbulator. I have sig generator Marconi TF995B. [200 KHz - 220 Mhz] Frequency counter Insteck SFG 2104 [1hz - 4Mhz] and Scope GOS 652G [to 50Mhz] If a Wbbulator is mandatory does anyone have one for purchase? Regards to all Ian Rose |
31st Oct 2017, 3:52 am | #2 |
Pentode
Join Date: Mar 2016
Location: Champaign, Illinois, USA.
Posts: 227
|
Re: Racal RA17W.
I tried various methods. The problem is measuring the output of the filter.
You need a signal in at TP2 from a 50 ohm source (through a say .001 uF cap) to swamp any reactance there. The loss in the filter is high. Thus measuring at TP3 is difficult. I used a well shielded line to my Airpsy SDR coupled in from TP3 through the stray capacitance to a short 1/4 inch piece of coax center conductor about 1/8 inch away. Any attempt at a direct connection bothers the tuning. I used a 6 dB attenuator to make sure the line was properly terminated. I just used a hand-tuned signal generator. The SDR tells both intensity and frequency. You could also use one of those DVB-T SDR dongles cheap on the Internet IF you've never used one for DVB (You have to make sure the driver provided for it has never been installed, and follow web instructions on loading the driver that hacks it to go down to 28 MHz. ) SDR# apparently still works with these. Both my filters had been hacked (bad mica caps in the distant past) and were way off. I finally got them tuned using the method of doing the coils one at a time, measuring each using a grid-dip meter with both adjacent coils shorted. You set all to exactly the same frequency .. which is not the center frequency! . Only the two at the ends are then tweeked using that analyzer. A nightmare. I never got the either filter perfect, but the set is WELL within specs. Oh yes .. the set needs to be well warmed up before adjusting the 37.5 MHZ filter. |
31st Oct 2017, 7:02 am | #3 |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
|
Re: Racal RA17W.
The ideal tool for tuning these filters is either a spectrum analyser with a tracking generator, or a vector network analyser. These were staggeringly expensive instruments when new, but some people in the group have acquired them on the second hand market and may be able to help.
What makes them more useful than a wobbulator and oscilloscope is that they are extremely sensitive in their receiver section. They need care to protect them from the voltages in an RA17, by using a resistive attenuating probe for the input to their receiver... IE two resistors and a blocking capacitor. And then they still have enough sensitivity to sniff the signal with very light coupling. A different attenuator and blocking capacitor is also needed for protection of the generator output. Dtvmcdonald is exactly right in what he says about coupling test gear to the filter. If you have to couple something other than very lightly, it pulls the tuning of the end stages of the filter and you happily adjust the filter to spec, then when you remove the connections, the filter tuning changes. Using a software defined radio dongle, a spectrum analyser or a VNA gives you the sensitivity needed to work with very light coupling. A luxury the military service centres never had. Racal had special jigs and purpose-built test gear back at the factory. There's a lot of adjustments and they all interact like crazy. The technique of tuning one stage at a time with its neighbours temporarily shorted is a powerful technique for killing the interactions and makes the adjustment task a lot less difficult. Just twiddling trimmers while watching a real time plot of the response is hard. The interactions and the number of adjustments will lead you astray unless you are very disciplined in your approach. The interactions tend to make your adjustment attempts diverge. For the day-job I've designed more complex filters than the RA17 ones, had to line-up my prototypes and write adjustment procedures for the production line and service centres. Filter adjustment is a major undertaking on anything other than basic designs, and the adjustment process costs the manufacturer a lot more than the filter parts did. Ideally, you need the help of someone who's done it before and has the test gear. Illinois and Fife are both a bit too far for house-calls. Is there anyone nearer? David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
31st Oct 2017, 12:47 pm | #4 |
Pentode
Join Date: Oct 2005
Location: Hebden Bridge, West Yorkshire, UK.
Posts: 199
|
Re: Racal RA17W.
Hi,
I don't think the "W" suffix has any particular significance, other than denoting yet another variation of the basic RA17. The main difference when comparing the W with the L version, lies in the IF bandwidths which are offered. The differences are as follows: W Posn.1 200Hz Posn.2 500Hz Posn.3 1.2kHz Posn.4 2.0kHz Posn.5 4.0kHz Posn.6 13.0kHz L Posn.1 100Hz Posn.2 300Hz Posn.3 1.2kHz Posn.4 3.0kHz Posn.5 6.5kHz posn.6 13.0Khz The W version is generally believed to have been made for the UK Meteorological Office for use with a specific RTTY modem, though I have never seen any positive confirmation of this. Not many were made so this model is quite rare compared to other versions. There is a Yahoo group - Racal Ra17 forum - where there has been much discussion of the various methods of aligning the 40MHz and 37.5MHz filters. You may find it useful to join that group and browse the postings on the subject. cheers Peter G8BBZ
__________________
Peter G8BBZ |
31st Oct 2017, 4:17 pm | #5 |
Nonode
Join Date: Nov 2015
Location: Welwyn Garden City, Hertfordshire, UK.
Posts: 2,013
|
Re: Racal RA17W.
I'm tempted to try a loosely coupled TV DSP dongle + PC with HDSDR or similar on those filters one day. Maybe in conjunction with a noise source.
TV Dongle method worked very well on aligning a mis-tracked HRO coil box. If my RA117 wasn't so heavy I would have tried this by now...it's well overdue. |
31st Oct 2017, 6:42 pm | #6 |
Pentode
Join Date: Aug 2010
Location: Isle of Wight, UK.
Posts: 217
|
Re: Racal RA17W.
Very many thanks to Mcdonald, David, Peter and Jon for such comprehensive comments and sharing of experiences on the subject.
Much to ponder and experiment with. Regards Ian |